DE102006015714B4 - Light-assisted testing of an opto-electronic module - Google Patents

Abstract

A test device for detecting defects of silicon and / or amorphous silicon in an optoelectronic module (10), comprising:
a. a first source (11) for generating an electromagnetic beam or particle beam (15);
b. a second source (12) for illuminating the optoelectronic module; and
c. a detector (13; 13a, 13b, 13c; 13d, 13e),
d. Means for terminating the illumination of the optoelectronic module prior to directing the electromagnetic beam or particle beam, the means being adapted to irradiate the optoelectronic module prior to the measurement of defects for an illumination duration of between 100 μs and 0.5 s.

Description

The present invention relates to a method for testing an optoelectronic module and to an apparatus for testing an optoelectronic module. In particular, the present invention relates to a method for detecting defective elements of an optoelectronic module and to a device for detecting defective elements of an optoelectronic module. The optoelectronic module has in particular the shape of a display element.

As the demand for non-CRT screen elements increases, the demands for liquid crystal display (LCD) and other display elements using switching elements such as thin film transistors (TFTs) are growing. In these display elements, the pixels are arranged in a matrix. For the purposes of the present application, "pixel" is understood to mean the complete RGB pixel, which is typically composed of three pixels. In this case, each of the three pixels is responsible for one of the three primary colors, namely red, green and blue. In the context of the present application, the term "pixel" is understood to mean a unit of an optoelectronic module which comprises a thin-film transistor, a pair of electrodes and a capacitor.

The switching elements of each pixel are typically connected via control lines, i. Gate lines and data lines activated. In order to ensure a good image quality of the display elements, none or only a very few of, for example, several million pixels may be defective. In order to ensure cost-effective production, it is therefore important, above all for the increasingly large display elements, to provide powerful online test methods. In these tests, the individual pixels are often tested with a corpuscular beam. The corpuscular beam can either be used to detect the charge applied via the leads and / or to apply charge to a pixel electrode.

The testing of the optoelectronic modules often takes place in such a way that a certain voltage pattern is applied to all pixels via the gate and data lines. Subsequently, for example, the individual pixels are irradiated by means of an electron beam and the resulting secondary electrons are measured. The measurement is supported by a counter potential arranged between the surface to be tested and the detector. that allows. to filter low-energy secondary electrons and thus to obtain information about the energy distribution of the secondary electrons. Depending on the voltage applied to the pixels, there is a typical energy range of the secondary electrons. Greater deviations hereof mean that the driven pixel has a defect that, for example, causes the secondary electrons emitted from that pixel to be too slow to overcome the opposing field applied in front of the detector. As part of a reporting system, the defect and the position of the defect are saved. Depending on the optoelectronic module, the defect can be repaired based on this information.

The EP 0 943 951 A1 teaches an apparatus and method for controlling optical surfaces. The proposed device comprises a DC light source for irradiating the entire surface of a LCD substrate with DC light. During a point defect measurement, a polarization shutter is opened to illuminate the substrate. In this case, a defect is made visible due to an amorphous silicon residue.

document US 5 057 773 A discloses a method for detecting defects by means of an electron beam. The electron beam is directed onto the sample, producing secondary electrons on the sample. Connections that are not at a given potential can be detected.

Also document US 2004/0223140 A1 teaches an apparatus for testing by means of an electron beam. A TFT display is tested in a two-step process: First, a coarse scan with an expanded beam takes place, after which a precise analysis takes place for a region detected as defective. The entire test time can be reduced.

The WO 2004/109 375 A1 teaches a method of testing substrates wherein a common terminal is formed on the substrate to short-circuit at least a portion of the wiring in a first area with at least a portion of the wiring in a second area. An electrical signal is applied to both areas and an electron beam is used to inspect the pixel electrodes with the help of the secondary electrons.

Also the EP 0 523 594 A1 teaches a method for corpuscular testing of liquid crystal display (LCD) substrates. In the case of a substrate for an LCD, potentials or currents are adjusted either by means of a corpuscular beam and / or potentials are measured and checked by detection of secondary electrons at different switching states of switching elements of the substrate.

It has been found that certain types of errors can not be detected by prior art testing methods, such as the test method described in the previous paragraph. An example of such a defect is amorphous silicon, which, for example, has not been completely removed in a lithographic step from areas which should actually be completely unmasked during the etching. During normal operation of the optoelectronic module to be tested, the amorphous silicon defect produced in the lithography step can result in the corresponding pixel being short-circuited and thus always producing the same polarization independently of the applied signals on the gate and data lines. The pixel is so defective and would have to be repaired. If too many pixels are defective, it is no longer economical to repair the optoelectronic module to be tested. It is therefore an object of the present invention to provide a method and an apparatus for testing optoelectronic modules, which overcomes the problems in the prior art. In particular, it is an object of the present invention to provide a method and a device for testing optoelectronic modules, wherein the method and the device are suitable for finding errors in the optoelectronic modules to be tested, which can not be found or can only be found incompletely in the prior art were.

The object is at least partially solved by the devices according to claim 1 and the method according to claim 32. Further advantages, features, aspects and details of the invention are evident from the dependent claims, the description and the accompanying drawings.

According to the invention there is provided an apparatus for testing an optoelectronic module having a first source for generating an electromagnetic beam or particle beam, a second source for illuminating the optoelectronic module and a detector.

Typically, the first source is for retrieving test results and the second source is for measuring a defect. In other words, this means that the measurement as such is made via the radiation or the particles of the first source, while the second source causes a change in the material to be measured, so that defective pixels become measurable, which without the radiation of the second Source would not distinguish from defect-free pixels. Typically, the second source is shaped and positioned such that the illumination incident on the opto-electronic module within a region comprising a plurality of pixels of the opto-electronic module has a substantially homogeneous intensity. Essentially in this context means that deviations below 15%. typically below 10%. The plurality of pixels is for example between 50x50 and 1000x 1000 such. 500x500. The area. within which homogeneous illumination takes place is typically the test area in which all pixels can be checked by deflecting the beam from the first source. For example, the area may range between 200mm x 200mm to 600mm x 600mm.

The second source typically includes at least one LED. An LED is a light emitting diode. The LEDs can be equidistant from each other.

The opto-electrical module is typically a module that is an element of a (color) screen and can be used as a screen for personal computers, portable computers, televisions, etc. Liquid crystals and color filters are usually not included in the optoelectronic module. Typically, the transistors are so-called thin film transistors (TFTs). It is typical that one or more thin film transistors and one or more pairs of electrodes are provided per pixel in a finished screen. Furthermore, one or more capacitors can be provided per pixel. The capacitor normally serves to prevent the voltage on the liquid crystal from dropping immediately when the corresponding transistor is turned off. Typically, the entirety of all pixels, including liquid crystals, form an LCD. Other elements of the LCD can be color filters as well as the screen cover plate. In typical embodiments, the optoelectronic module is the bottom plate of a screen, the bottom plate comprising a plurality of thin film transistors, a plurality of electrodes, and a plurality of capacitors. Liquid crystals are typically not included in the optoelectronic module to be tested according to the invention.

The second source is typically located between the opto-electronic module under test and the first source. In addition, embodiments are possible in which the first source is arranged between the optoelectronic module and the second source, or in which the two sources are mounted at a height. It is also conceivable that the two sources are mounted in a common holder.

Typically, large optoelectronic modules are tested such that the apparatus for testing comprises at least two first sources, at least two second sources, and at least two detectors. In other words, it means that parallel testing is possible. It can be found Applications containing at least 10,000 pixels on the optoelectronic module. Typically, the optoelectronic module to be tested according to the invention comprises at least one million pixels.

1. a. Beleuchten des optoelektronischen Moduls;
2. b. Richten eines elektromagnetischen Strahls oder Teilchenstrahls; und
3. c. Detektieren von Defekten in dem optoelektronischen Modul.

According to another aspect of the present invention, there is provided a method of testing an optoelectronic module comprising:

1. a. Illuminating the optoelectronic module;
2. b. Directing an electromagnetic beam or particle beam; and
3. c. Detecting defects in the optoelectronic module.

By illuminating additional to directing the electromagnetic beam or particle beam, defects are visualized that would not be detectable without illumination.

Typically, the detection takes place only within a period of time in which a voltage applied to a defect-free thin-film transistor of the optoelectronic module has dropped to a maximum of 80% or 60%. The electromagnetic beam or particle beam is typically directed to the opto-electronic module or to a detector unit. Typically, the voltage is measured on at least one pixel of the optoelectronic module. Based on the voltage across a plurality of pixels, an average voltage value may be calculated and the measurement of the voltage of each pixel compared with the average voltage value. A pixel is classified as defective if the measured voltage value deviates by more than a limiting percentage from the average voltage value. Typical limit percentages are between 20% and 40%, especially between 25% and 35%, such as 30%. The illumination preceding the end of the measurement should typically take place only for a time interval in which the voltage across a defect-free thin-film transistor of the optoelectronic module has not fallen by more than 20% or 30%, at most 50%. The lighting is completed when the measuring starts.

The testing can take place in a vacuum chamber. The device according to the invention may contain one or a plurality of vacuum chambers. The vacuum chambers of the plurality of vacuum chambers may typically provide vacuums of different levels. This means that at least the beams of the first source are guided in a vacuum chamber. It is typical that the detector is also located in the vacuum chamber. Alternatively, it is also possible to test in an open environment. In particular, the light of the second source may be room light.

• 1: Beispielhaftes Pixel aus einem optoelektrischen Modul:
• 2: Beispielhaftes fehlerhaftes Pixel aus einem optoelektrischen Modul;
• 3: Eine erste Ausführungsform der vorliegenden Erfindung;
• 4: Eine zweite Ausführungsform der vorliegenden Erfindung;
• 5: Eine dritte Ausführungsform der vorliegenden Erfindung;
• 6, 7: Zwei mögliche Ausführungsformen der zweiten Quelle;
• 8: Einen Querschnitt durch einen Ausschnitt aus 6;
• 9: Zeit-Signal Diagramm zu einer Ausführungsform des erfindungsgemäßen Testens;
• 10,12:Darstellung des Spannungsabfalls in einem Defekt aus amorphen Silizium und einem Dünnschichttransistor bei kurzer bzw. anhaltender Beleuchtung; und
• 11: Zeit-Signal Diagramm bei anhaltender Beleuchtung
• 13: Messergebnisse in Abhängigkeit der dem Messen vorangegangenen Beleuchtungsdauer.

The invention will be explained by way of example with reference to the accompanying figures. Show it:

• 1 : Exemplary pixel from an optoelectric module:
• 2 : Exemplary defective pixel from an opto-electrical module;
• 3 A first embodiment of the present invention;
• 4 A second embodiment of the present invention;
• 5 A third embodiment of the present invention;
• 6 . 7 Two possible embodiments of the second source;
• 8th : A cross section through a section 6 ;
• 9 : Time-signal diagram for an embodiment of the testing according to the invention;
• 10 . 12 : Representation of the voltage drop in a defect of amorphous silicon and a thin-film transistor in short or continuous illumination; and
• 11 : Time Signal Diagram with Continuous Lighting
• 13 : Measurement results as a function of the illumination duration preceding the measurement.

1 shows an exemplary pixel of an optoelectronic module. The electrodes 104 are provided by logic (not shown) over the data line 101 and the gate line 102 driven. That is, if both a signal on the gate line 102 is applied to the thin-film transistor 103 on passage, as well as a data signal on the data line 101 is applied to the electrodes 104 of the pixel, which typically causes the molecules to align in a different direction than they were without voltage, thus rotating the polarization direction of the light passing through them by some angle. The liquid crystal and color filters are typically not included in the optoelectronic module which is tested according to the invention. The construction of 1 further includes the capacitor 105 , A capacitor connected to the electrodes of the liquid crystal is typically used so that the voltage applied by driving the illustrated cell does not immediately drop when the gate line and data line are no longer simultaneously Voltage applied. The capacitor 105 for example, with the gate line of an adjacent pixel or, as in 1 shown, with mass 106 be connected. By ground is meant in this context a common potential, to which the capacitors of all pixels are connected. This can be grounded. However, there are also embodiments in which it is not grounded, but is on a defined, typically adjustable value.

2 should be an exemplary defect 107 represent amorphous silicon, which can be found according to the invention. The optoelectronic module typically comprises silicon and / or amorphous silicon. The amorphous silicon is typically deposited to form the TFT, after which it is removed again in the regions outside the transistor, for example by a masking step and a subsequent etching step. It is possible that, for example due to contamination, the mask is damaged, or that the etching process as such is not carried out without errors, so that residues of the amorphous silicon remain outside of the TFT. These residues can lead to the malfunction of the pixel, z. Because they produce a short circuit in the event that they are illuminated. Frequently, optoelectronic modules, in particular LCD screens in normal operation, comprise a light source which illuminates all pixels from behind, ie from the side facing away from the user, with, for example, white polarized light. Depending on the voltage applied to the individual electrodes of the pixel, the polarization direction is rotated in the liquid crystal. In this case, the polarization direction can be rotated such that the light can not pass through the screen cover plate provided between the liquid crystal and the user, which is provided with a polarization filter. In this case, the corresponding pixel remains dark. However, if the pixel has a defect due to an amorphous silicon residue, this can cause a short circuit under light and the corresponding pixel always appears bright or always dark, regardless of the applied signal.

The production of the transistor 103 includes, among others, the deposition and etching of amorphous silicon. The region in which the amorphous silicon is provided, particularly within the thin film transistor, is masked, while all regions outside thereof are without a mask. A typical dry chemical etching step is performed. In the in 2 In the example shown, it happens that even after the etching step, the rest 107 of amorphous silicon left over from the TFTs. The amorphous silicon is non-conductive as long as no light falls on it. Will that be in 2 Therefore, pixels tested in the dark with the prior art methods can not detect an error, the amorphous silicon behaves as an insulator. However, it is typical for the use of the cell in an optoelectric module that light passes through the cell. As the light intensity increases, the amorphous silicon increasingly loses its insulating properties and gains in electrical conductivity. This triggers the in 2 illustrated defect 107 during operation of the optoelectronic module a short circuit between the electrode 104 and mass 106 so that the pixel shown is defective overall. The pixel testing according to the invention comprises the step of illuminating the pixel with a certain dose of light. This will be the amorphous silicon 107 conductive and leads in the example shown that during the test that over the data line 101 and gate line 102 applied charge attached to the electrode 104 of the pixel lies directly to the ground 106 can drain away. The defect 107 can be detected thereby (for the test procedure in detail see later).

3 shows a first embodiment of the device 1 for testing according to the present invention. The device comprises a first source 11 for generating an electromagnetic or particulate beam 15 , Typically, the generated beam is an electron beam; Also conceivable are ion beams. Alternatively, electromagnetic radiation, in particular from the visible spectrum, can be used for testing. Although reference is repeatedly made in this application to a device with an electron beam source, this is not to be understood as limiting. The generated beam 15 moves along the optical axis 14 , In general, it is possible to focus the beam between the source and the optoelectric module to be tested, in particular to focus, collimate, align, filter, accelerate, decelerate, deflect, and / or correct for astigmatism, etc. For this purpose, the device may be used for testing have one or more of the following: optical lens, magnetic lens, electrostatic lens, combined electrostatic magnetic lens, Wien filter, condenser, aligner, collimators, deflectors, accelerators, decelerators, diaphragms, stigmators, etc. In the case of particle beams it is it is possible for the particles after the source to be brought to a high speed, for example by being carried out by acceleration electrodes, and decelerated again shortly before they hit the optoelectrical module. This has the advantage that interactions between the particles that expand the beam can be reduced.

In 3 is an example of a deflector 16 shown, with the help of which the beam can be deflected to different points on the optoelectric module 10 to drive with the beam. In the case of a particle beam, the deflector may generally be one or more electrodes or magnetic deflection coils. In the case of an electromagnetic beam, for example, one or more mirrors or prisms may be used. A deflector is not absolutely necessary if the control of different points on the optoelectronic module via an exact movement system of the stage 9 can take place, wherein the movement system, the opto-electric module 10 typically move in the xy plane. The xy plane is the plane that is substantially perpendicular to the optical axis of the device for testing. The optical axis is typically defined by the direction of the beam of the first source. The testing of all cells of the optoelectrical module can generally also be carried out in such a way that the surface to be tested is first divided into a plurality of test areas. With the help of one or more deflectors, all the cells within a test area are irradiated and tested in turn, whereby the movement system of the stage is not moved during this period. When the test area has been completely checked and the results stored, the motion system shifts the opto-electrical module so that all cells of a new test area can be irradiated and tested. In this way, the complete opto-electrical module can be tested. In this regard, it is typical, particularly in view of the growing sizes of opto-electrical modules, that multiple inventive devices for testing an opto-electrical module are used in parallel. In other words, in this case, the device according to the invention for testing an optoelectronic module comprises at least two first sources, at least two second sources, and at least two detection units. The at least two devices for testing tested in parallel test different, typically adjacent, test areas in test mode. The total test time for testing an opto-electronic module in the case of n parallel-operated devices for testing is reduced to about 1 / time that would be needed in the case of only one device used for testing. The typical size of the test areas is between 20 × 20-40 × 40 cm ² , in particular around 30 × 30 cm ² . Typically, at least one of the deflectors used is able to provide in each direction a deflection of the beam of +/- 10-20 cm, in particular +/- 15 cm.

The beam 15 in 3 is applied to a cell of the optoelectronic module to be tested 10 directed. The beam impinging on the optoelectronic module 15 causes secondary particles 17 arise and leave the optoelectronic module. The number of secondary particles and the amount of their energy allow conclusions about defects within the irradiated cell of the optoelectronic module. The secondary particles are with the detector 13 measured. Typically, the detector comprises a scintillator, a photomultiplier, and further units for reading out and evaluating the information obtained. It is also typical that a counterpotential can be applied shortly before the detector. For this it is typical that, for example, a grid or an electrode ring on which a potential can be applied, in front of the detector 13 is arranged. Depending on the applied voltage, all secondary particles are filtered that can not overcome the applied potential. These do not cause a rash in the detector. Alternatively, energy filtering may also be performed by means of a magnetic field or a combination of a magnetic field and an electric field. reference numeral 12 represents an annular second source, typically the optoelectronic module 10 illuminated with red or green light. It is generally preferred that the source be shaped and positioned such that the light intensity on the optoelectronic module is constant or nearly constant within a complete test area. This creates identical test conditions for all pixels within the test area. In particular, the TFT and the defects, in particular the defects of amorphous silicon, are thereby illuminated with the same light intensity. The conductivity of the TFT thus increases in all pixels in the same or almost the same way and the conductivity of the defects of amorphous silicon thus increases in all pixels in each case in the same or almost the same way. The position of the second source may be generally above the optoelectronic module or below the optoelectronic module. ie on the side of the radiation source or on the opposite side of the optoelectronic module. In the latter case, it is typical for the display to be on a stage which is transparent to the wavelength of the second source. Furthermore, the second source may also be stray light from the environment.

The second source typically provides red light. The light may for example have a wavelength between 550 and 800 nm, such as. B. 630 nm. Alternatively or additionally, green light from the second source can be provided. Both red and green light is useful to substantially improve the conductivity of amorphous silicon. Typically, the detection of the secondary particles is done with a detector unit comprising a scintillator and a photomultiplier. The influence of stray light and reflected light reaching the detector on the photomultiplier can be reduced when red light is generated at the second source.

The optoelectronic module typically has contact elements (so-called contact pads), via which an electrical contact to a test unit can be produced. The device according to the invention typically allows the beam of the first source to be directed to individual pixels of the optoelectronic module. This means that only a negligible part of the beam falls on adjacent pixels. Insignificant here is understood to mean a proportion of not more than 20%.

4 Fig. 10 is an illustration of a second embodiment of an apparatus for testing the present invention. In addition to the already off 3 known elements, numerous further elements are exemplified in this embodiment. However, it is emphasized that this is merely an illustrative representation, and the elements are 4 in no way mandatory for the execution of the invention are required. For carrying out the invention, only the features specified in the independent claims of the present application are required.

4 shows how 3 an apparatus for testing in which the first source is a particle beam source. The first source 11 that is a cathode 11a and an anode 11b includes, generates a beam 15 of particles, for example of electrons. The anode can simultaneously act as a shutter. In front of the cathode 11a can be a grid 11c be arranged, which lies at a certain potential. The beam initially moves in a channel 27 , the so-called "liner tube" on the optoelectronic module 10 to. Between anode and optoelectronic module are shown by way of example a condenser lens 21 , two more lenses 22 and 23 which represent a projective and focus lens, the stigmator 24 , the electrostatic deflector 28 and the magnetic deflector 16 , The stigmator is used to correct astigmatic errors in the beam. A deflector is typically used for a fine deflection, another deflector for a rough steering. In the present example, the electrostatic deflector provides the fine deflection, while the magnetic deflector causes the main deflection. The fine deflection can be made very promptly due to the use of an electrostatic deflector. The second light source 12 is illustrated as a ring of LEDs (light emitting diodes), the ring being generally and not limited to the present embodiment such that the optical axis forms the ring center. Just before the detector is a conductive grid 25 arranged, to which a voltage can be applied. In general, any type of spectrometer or energy filter may be arranged instead of the grating. In other words, the device according to the invention for testing optoelectronic modules may comprise means for energy filtering the secondary particles. The detector off 4 consists of a scintillator 13a , a photomultiplier 13b and a light detector 13c , Not shown in 4 is the logic for evaluating the signals received in the detector. Such logic is typically connected to the test unit (not shown), which applies a certain voltage pattern to the optoelectronic module, as well as the means for energy filtering the secondary particles, such as e.g. B. the grid 25 in 4 ,

4 also shows stepper motors 26 for mechanical alignment of the cathode and the vacuum pumping system 20 , which is responsible for creating a vacuum within the device for testing. It is possible that the pumping system has a plurality of exhaust nozzles attached to the device according to the invention for testing. In the example of 4 the pumping system has two suction nozzles. Moreover, it is also possible that vacuums of different strengths are produced in different chambers of the device according to the invention. The optoelectronic module is connected to the test unit 18 connected, which applies voltage to the data and gate lines of the optoelectronic module according to the test pattern used. Not shown in 4 is the stage.

Furthermore, in 4 octopole plates 29a and 29b shown. The octopole plates 29b are provided with a static potential for generating a suction field. The secondary electrons are thereby accelerated upwards. This is also known by the term "collecting" of the secondary particles. The second source 12 may be shaped and mounted so as to aid in collection in the event of a voltage applied thereto. The octopole plates 29a are controlled dynamically depending on the primary beam. They serve the secondary electrons to the detector 13a-c to accelerate. It should be emphasized, however, that these octopole plates are not necessarily required for carrying out the testing according to the invention. The octopole plates 29b illustrate exemplary embodiments of a static detection unit and the octopole panels 29a illustrate exemplary embodiments of a dynamic detection unit. Such detection units may be included in embodiments of the device according to the invention in general.

5 shows a further embodiment of the present invention. In the embodiment according to 5 becomes the optoelectronic module to be tested 10 from below with the light of a second source 12 illuminated. From above, the one from the source 11 outgoing electromagnetic radiation the lower detector 13d sent, for example, a deflection of the beam direction by means of the prism 19 is shown. The radiation of the first source may, for example, be light from the visible spectrum. The detector 13d e tests the optoelectronic module for defects. The lower detector 13d can z. B. crystals, which react differently depending on the applied electric field strength. In this case, the detector is placed in close proximity to the opto-electronic module for testing, with the distance between the opto-electronic module and detector typically on the order of 10 ¹ μm, such as 10 μm. B. 30 microns, is. Depending on the applied voltage, the crystals react differently. This different reaction can be over that of the lower detector 13d returned light from the first source at an upper detector 13e be read. The light of the source 12 can also be the ambient light of the room.

6 and 7 show two exemplary embodiments of the second source 12 , 6 shows an arrangement of LEDs 120 that the shape of a ring 121 having. 7 shows an arrangement of LEDs 120 that form the shape of an inner ring 121 and an outer ring 122 having. Typically, the second source includes a plurality of individual light spot sources. A typical light source is an LED. In addition, it is typical that the light spot sources represent the end pieces of a glass fiber connection to a common light source. It is also typical to use as the second source a flat light source, such as a conventional light bulb or tube. Furthermore, filters, in particular color filters, or diffusers can additionally be attached.

The use of LEDs has the advantage that light of known and substantially uniform wavelength is generated. However, the light from LEDs is also directed. Therefore, it is typical for a plurality of LEDs to be arranged in such a way that a light intensity which is as uniform as possible in terms of strength is formed on the region of the optoelectronic module to be tested. Typically, more than 20, 25 or 50 LEDs are arranged. For example, 80 or 100 LEDs may be arranged on a ring. The orientation of the LEDs is directed substantially perpendicular to the optoelectronic module. Essentially in this case means that deviations in the range of up to 20 ° to 30 ° are possible. Thus it has been shown in experiments that a slightly directed arrangement allows a very uniform intensity distribution on the area of the optoelectronic module to be tested. The LEDs were in a ring according to 6 or 7 arranged, wherein their orientation was carried out at an angle of 14 ° to the optical axis of the device according to the invention. In general, alignment angles between 0 ° and 25 ° give good results, especially angles between 10 ° and 20 °. In 8th By way of example, a section of a second source is shown. wherein the cross section is shown through a side of the annularly arranged plurality of LEDs. The LED shown 120 is in the ring in a recess 31 attached. The depression has walls 34 on that at a defined angle from the root 32 lead the LED away. This will cause the beam area 33 the LED 120 already restricted to a defined angle range. Typically, this points through the walls 34 limited beam area 33 an opening angle of between 20 ° and 90 °, often between 30 ° and 60 ° such as 45 °. The LED, in general and not limited to the present embodiment, is typically located in the recess such that the head tip of the LED is still within the recess and does not look beyond it. Typically, the distance 35 between head tip of the LED and the edge of the second source a few mm, z. B. between 1/4 mm and 4 mm, in particular 1 mm. Furthermore, the LED is at an angle of 14 ° to the optical axis 14 aligned with the device according to the invention for testing. The alignment will be through the line with the reference number 30 displayed. The optical axis 14 that is at the center of the annular second source 12 is located in 8th also marked. With 14 and with 30 marked lines 8th have an angle of about 14 °. Angles of 2 °, 5 °, 10 °, 13 ° and 17 ° also lead to a good homogeneity in the intensity distribution. In an annular and equidistant arrangement of about 80 LEDs, the ring has a diameter of about 40 cm, and at an orientation of about 14 ° to the optical axis and a distance of about 23 cm between the annular second source 12 and the surface of the optoelectronic module to be tested, an excellent, almost homogeneous light intensity can be generated on the optoelectronic module. It should be emphasized, however, that these are merely exemplary values, and there are a variety of other ways to produce a high degree of intensity homogeneity in the area of the optoelectronic module under test.

The outer material of the LEDs used is typically an insulator. In order to avoid that in particular backscattered electrons accumulate on the LEDs and thus may possibly lead to disturbing fields, it is possible to attach means for discharging charges in front of the LED. These can in particular be a conductive network. The material in which the LEDs are mechanically anchored is typically a conductor. As a result, not only charged secondary particles and particles scattered back from the optoelectronic module can flow away, but the second source can to some extent Potential that supports the collection of secondary particles (see the description 4 ). Typical potentials of the second source are between 0 and -100 V.

The 9 and 11 show the time course of two embodiments of testing, according to the in 9 a short pulse of light precedes the measurement and according to the described in 11 example described a constant lighting takes place. The 10 and 12 show to the embodiments 9 and 11 in each case, for example, the voltage profile at a defect made of amorphous silicon and at a defect-free thin-film transistor. The light of the second source used may in particular be all the light sources described so far. The test unit is a circuit known in the prior art, at whose outputs a plurality of signals of different heights can be tapped at different times. The test unit may be, for example, a computer having an input unit, such as an input unit. As a mouse and / or a keyboard, a presentation unit such. B. a screen, a computing unit, such. B. a CPU (central processing unit), and a memory unit. such as B. a non-volatile memory, for example. A hard disk, and / or a volatile memory such. B. a random access memory (RAM) has.

9 shows according to the invention the time course of a test procedure. By means of the test unit is at the time t0 a voltage is applied to the data line. This is usually done just before the gate at the time t1 is opened. Between t1 and t2 Both the gate is open and a signal on the data line. This process is also called the "bustle". At the time t2 For example, the gate is again turned off by, for example, being reset to a zero potential, or, as in FIG 9 shown to a negative potential. The latter has the advantage that defects within the gate such. B. a short circuit between the gate and data line can be detected. In general, the gate is disabled when the applied voltage drops below the voltage on the data line. The signal on the data line is typically only after that - namely at the time t3 - changed. The bustle is complete. At the time t3 The data signal can either be set to 0 or switched to a different value than it is between t0 and t3 present. The latter, in turn, has the advantage that a defect in the transistor, which leads to the fact that, in spite of the actually blocked transistor, this allows the changed data signal to pass through, can be detected. After a short wait between t3 and t4 is in the embodiment of 9 the optoelectronic module with a controlled, short light pulse of length t5 - t4 illuminated. Typical light doses of such a light pulse are between 5 lx · s (lux * seconds) and 30 lx · s, in particular between 10 lx · s and 20 lx · s such. B. 13 lx · s. The dose of light used is typically achieved with a light of between 500 lx and 2500 lx illuminance ( 1 lx = 1 lm / m ² ), in particular between 1000 lx and 1500 lx such. B. 1300 lx has. The term "illuminance" refers to the illuminance in the plane of the optoelectronic module and not to the illuminance at the second source itself. For example, the illumination duration can be selected between 0.001s and 0.1s. B. 0.01 s. Thus, if the second source is operated for a period of 0.01 s with an illuminance of 1300 lx, the result is a light dose of 13 lx.s. The illumination period is typically chosen such that the amorphous silicon of the defects becomes conductive for a sufficient period of time. The period of time is sufficient if the potential previously applied in the pixel can flow away via the defect at a rate such that the voltage drop can be detected in the subsequent measurement. This typically means that at least 20% or 30% of the charge in the defective pixel should have already drained off during the measurement. At the same time, the illumination period prior to the measurement may be determined in accordance with the procedure given in 9 described embodiment are not chosen too long, because then the charge would drain significantly over the amorphous silicon of the defect-free transistor and thus a defective pixel would not be distinguishable from a defect-free transistor. The waiting time between t4 and t3 can also be omitted in general, then t3 = t4. The advantage of a waiting time, however, is that the voltage after switching the data line at the time t3 stabilized a bit before then at t4 the lighting is switched on. Typical waiting times for the time interval between completion of the act ( t3 ) and lighting ( t4 ) are between 50-100 μs.

If the light source as in 9 is turned off during the measurement, the measurement is not subject to a time pressure that goes beyond the known time limits of the prior art. Since the voltage gradually drops even in the defect-free pixels even without illumination, at some point in time t *> t6 so much charge has flowed out of all the pixels that meaningful measurement results can no longer be achieved. The time t * - t6 is typically on the order of 10 ⁰ s, z. B. between 0.5 s and 5 s. In order to continue the testing of the optoelectronic module, a so-called "refresh" must now take place. This means that, unless the method of testing is yet to be completed, the description and lighting procedure, e.g. In 9 shown, restarted. That is, first the data signal is opened again (time t0 ), then a gate voltage applied ( t1 ) etc. The entire procedure is repeated until all the pixels of the area to be tested of the optoelectronic module or of the complete optoelectronic module have been examined for defects.

In 10 shows the voltage drop across a defect-free thin-film transistor ("TFT") and a defect of amorphous silicon ("a-Si defect") during and after illumination. Initially the voltage is at the TFT and at the a-Si defect at V0. This corresponds to the tension that has set after the drive. Without light the tension decreases only very slowly, in which 10 this scale is barely recognizable. Will be at the time t4 the lighting is switched on, until the time t5 , ie as long as the illumination is directed to the optoelectronic module, a significant voltage drop in the a-Si defect takes place. A voltage drop can also be seen in the transistor, but this is much smaller than the voltage drop in the defect made of amorphous silicon. After switching off the lighting, ie after t5 , the voltage in the TFT and in the a-Si defect drops only very slowly. The slow drop is due to the always present leakage currents. For the measurement, this means that the voltage in a pixel with an a-Si defect from the time t5 a good distinction is the voltage in a pixel in which there is no defect, and the voltage drop is caused only by the defect-free transistor. At a later date, a few seconds later t5 or. t6 can lie. Due to the known and unavoidable leakage currents in the TFT, the voltage in the TFT has likewise dropped to such a low value that the difference to the voltage in a pixel with a defect of amorphous silicon is no longer sufficiently clear. In this case, to continue the measurement, a refresh must take place. The timing of the measurement t6 can generally with the timing of the completion of lighting t5 coincide. Alternatively, it is possible to wait a short time between the completion of the illumination and the beginning of the measurement. Furthermore, it is possible to start the measurement already before the lighting is finished. In this regard, however, care must be taken that the stress in a typical a-Si defect at the beginning of the measurement should already have dropped sufficiently in comparison with the voltage at the defect-free TFT.

11 shows a modification of the in 9 described method, which consists in that instead of a short pulse of light, the optoelectronic module is constantly illuminated by the second source. For the blowing process, this means in comparison to the method 9 no or no significant difference. At the time t3 is the pixel driven and is at a certain voltage. The light from the second source illuminates the pixel so that the amorphous silicon is conductive and voltage can drain. As already mentioned, defects of amorphous silicon become conductive and lead to a drop in the voltage. In addition, although to a lesser extent, the increased conductivity of the amorphous silicon in the transistor caused by the light also leads to a voltage drop across the transistor. These two circumstances should be considered when choosing the dates t6 and t7 be taken into account. Because: between t3 and t7 there must be a sufficiently long period of time for voltage to drop across the defects of amorphous silicon. This means that it is not possible to start measuring directly after the measurement, because in this case some defects of amorphous silicon could not be visualized yet, because the voltage takes a certain amount of time to decay. At the same time, the time interval between the end of the process ( t3 ) and the beginning of the measurement ( t6 ) are not too large, since in this case voltage would have dropped not only in the defective pixels but also in the defect-free pixels, namely via the amorphous silicon in the thin-film transistors. In general, defect-free pixels should not have lost more than a maximum of 10% or a maximum of 20-30% of the originally applied voltage by the end of the measurement process.

Furthermore, the measurement period in the example is off 11 by the time t7 limited shown. This results from the fact that the light of the second source remains switched on during the entire measurement. The amorphous silicon in the transistors is therefore conductive during the measurement so that the voltage of the pixels can flow continuously through the (defect-free) transistors. At the time t7 or shortly thereafter, it is no longer possible to distinguish in the measurement, whether the measured voltage drop was actually caused by a defect or by the defect-free transistors. At this time, so the refresh must take place, and the entire procedure of, for example. 11 be repeated. The period of time that can be measured in the case of a second source that remains lit is generally much less than the time period available to measure when the illumination is off during the measurement period. Under much smaller here time differences in the amount of at least one order of magnitude understood. The period available for measurement according to the in 11 example described is, for example, a maximum of 50-80 ms. If the measurement continues after that, it is hardly possible to distinguish defective from defect-free pixels. The on 9 Test method according to the invention, which only irradiates the optoelectronic module with a short dose of light before the measurement, therefore, generally requires fewer refresh cycles than the test procedure, which involves constant illumination. In addition, greater waiting times between driving ( t3 ) and measurement ( t6 ) to get voted. This has the advantage that even defects that only lead to a slow voltage drop can be detected.

12 shows the voltage drop in the TFT and in the a-Si defect in the example of FIG 11 in that the lighting is switched on before, during and after the measurement. The voltage applied to the TFT and the a-Si defect after the drive V0 In the a-Si defect due to the illumination in a relatively short time clearly decreases, while the decrease takes place in the TFT slower. Due to the continuous lighting, however, the voltage drop in the TFT is much more pronounced compared to the situation in which the lighting is switched off again after a short time. The time window in which a measurement can be performed is given by [ t6 ; t7 ] Are defined. In this time interval, the voltage at a defect-free pixel is still sufficiently different from the voltage at a pixel that has a defect of amorphous silicon. To t7 However, the voltage drop in a defect-free pixel due to the leakage current in the TFT, which is significantly more pronounced by the illumination than the leakage current through an unlit TFT, so strong that defective and defect-free pixels can not be distinguished with sufficient certainty. Before the testing is continued, a refresh must take place.

The measurement will be explained below by way of example using an electron beam microscope. First, for example, the pixels are named after one of the 9 and 11 set to a certain voltage. Typically, all pixels are set to the same voltage, such. B. +/- 5 V or +/- 15 V. Alternatively, it is also possible that the pixels are alternately occupied with a positive and a negative voltage. In this case, additional defects can be found due to which leakage currents occur between two adjacent pixels. For example, all even-numbered pixels in a row can be driven to + 10V, while all odd-numbered pixels in the row are driven to -10V. In this case, it makes sense to set all odd-numbered pixels to -10 V in the adjacent row and all even-numbered pixels to +10 V. This ensures that the four closest neighbors of each pixel have an opposite voltage than the pixel itself Has.

The optoelectronic module is - depending on the size and deflection possibilities within the electron microscope - divided into several test areas to be tested. The movement system of the stage allows the optoelectronic module to be moved in such a way that the different test areas can be approached via the movement system. While testing a test area, the motion system is at rest; the pixels are controlled via the deflectors installed in the electron microscope. Typically, the beam may be deflected in both the x and y directions. The x-y plane is defined as a plane perpendicular to the optical axis of the electron beam microscope. The electron beam is directed to this pixel per pixel for a period of time. The secondary particles - typically secondary electrons - are measured, normally spectroscopic means or energy filters, such. As a verifiable with a potential grid, are provided in front of the detector. The secondary particles typically leave the opto-electronic module with energy composed of two components. The first component results from the energy distribution of emitted secondary particles typical for the material to be tested. The second component results from the voltage of the pixel. If this is negative, this leads to an increased energy of the secondary particles. If this is positive, the energies of the secondary particles are smaller than after the typical energy distribution of secondary electrons on the material to be tested. For example, if the voltage is 0 or close to 0 due to a defect, the total energy is substantially equal to the energy from the first component.

The evaluation of the measured data is typically carried out in a comparison algorithm for all pixels. If, for example, all the pixels are set to -15 V during the driving, and the measurement takes place so long that a maximum of 10% of the voltage has already dropped in the defect-free pixels, this means that the secondary electrons emitted by defect-free pixels, have at least 13.5 eV. Typically, they then have energies of up to about 25 eV, with the energies composed of the two components mentioned above. In this example, the secondary electrons on defect-free pixels, in addition to their typical energy distribution with energies up to 10 eV, receive an additional energy of 13.5 eV - 15 eV, which is caused by the applied negative voltage. Pixels in which, for example, due to an amorphous silicon defect, the voltage has fallen to, for example, 60% of the original voltage are defective and should be detected as such. In this example, at 60%, the resulting voltage in the defective pixel is -9V. For example, the grid placed in front of the detector is set to a voltage of -15V. This means that almost all secondary particles emitted by defect-free pixels reach the detector and are detected there. However, the secondary particles emitted from the defective pixel having only a voltage of -9V can not largely overcome the counterpotential of -15V. More specifically, only the secondary particles from the defective pixel can reach the detector which, due to their first energy component, has at least 6 eV. This leads to a significant difference in the detector results between non-detectable pixels and defective pixels.

Per pixel, the number of measured secondary electrons can be visualized. A high number can be represented as a bright point, a comparatively low number as a dark point. Comparatively low means compared to the number measured at the pixels of the vicinity or the entire area to be tested. 13 shows such an optical representation. The reference numbers 41 - 45 refer to different measurement results of a measurement performed on the same test area of an optoelectronic module, wherein the light dose, which was irradiated before measurement on the optoelectronic module, varied. In the with 41 only a small difference in brightness can be detected. The light dose radiated before the measurement was 0 lx.s, ie there was no illumination before the measurement. This made the defect of amorphous silicon nonconductive; he can not be detected. The light dose at the second measurement, the measurement result with the number 42 referred to was 6.5 lx.s. The defect 107 made of amorphous silicon is already visible, since significantly fewer secondary electrons could be measured here. The difference in the measurement result becomes even clearer 43 measured at 13 lxs. However, the contrast between defect-free and defective pixels begins to decline as the light dose continues to increase. This is how the measurement became with the measurement result 44 with a previous light dose of 32.5 lx.s and the measurement with the measurement result 45 with a previous light dose of 65 lx · s performed. The comparatively higher dose of light already causes the number of secondary electrons emitted by defect-free pixels also to decrease. The contrast between defect-free and defective pixels is reduced as light dose increases.

Typically, the number of secondary electrons emitted by a pixel is set in relation to an average value composed of the corresponding number of secondary electrons emitted from the adjacent pixels. Typically, small areas of, for example, 4 × 4, 8 × 8, or 10 × 10 pixels are used as a basis for calculating the average value. Thus, the comparison can always be done locally with the neighboring pixels. It is customary not to compare the measured numbers of pixels, but a normalized detector value. A normalized average detector value may be, for example, 120 in one embodiment. If the normalized detector value of a pixel deviates from a previously defined limit value, then this pixel is considered to be defective. Typical limits are between 20% and 40%, especially at 30%. In the present example, for example, a deviation of more than +/- 30% for a pixel is observed, ie. if the normalized detector value at this pixel is less than 120 * 0.7 = 84 or greater than 120 * 1.3 = 156, that pixel is classified as faulty. In other words, if the ratio of the normalized detector value of a pixel to the normalized average detector value is between 0.7 and 1.3, the corresponding pixel is classified as defect-free. If the ratio is below 0.7 or above 1.3, the pixel is classified as defective. This information is stored in each case.

If the optoelectronic module has been illuminated too long before the measurement, then already everywhere, i. even with the defect-free pixels, a significant drop in voltage occurred. The normalized average detector value is z. For example, at 70. The normalized detector value of the defective pixel is, for example, 60. The defective pixel can therefore no longer be classified as defective, since the difference to the defect-free pixels has become too small. The measurement can only be continued after another refresh.

The illustration of the measuring method was only an example to increase the clarity of the present application. This should by no means be understood as limiting. In general, voltages of all possible heights can and will be applied to the pixels and the spectroscopic means mounted in front of the detector. These can also differ by orders of magnitude of the voltage in the measuring example described above.

Claims (57)

A test device for detecting defects of silicon and / or amorphous silicon in an optoelectronic module (10), comprising: a. a first source (11) for generating an electromagnetic beam or particle beam (15); b. a second source (12) for illuminating the optoelectronic module; and c. a detector (13; 13a, 13b, 13c; 13d, 13e), d. Means terminating the illumination of the optoelectronic module prior to directing the electromagnetic beam or particle beam, the means being adapted to detect the optoelectronic module prior to the measurement of defects for a Illumination time between 100 μs and 0.5 s to be irradiated. Test device after Claim 1 wherein the second source is shaped and positioned such that the illumination incident on the optoelectronic module has a substantially homogeneous intensity within a region comprising a plurality of pixels of the optoelectronic module. Test device after Claim 1 or 2 , wherein the second source emits a wavelength of at most 800 nm. Device after Claim 1 . 2 or 3 wherein the first source is for generating test results and the second source is for measuring a defect (107). Apparatus according to any one of the preceding claims, wherein the beam of the first source can be directed to individual pixels (103, 104, 105) of the optoelectronic module. Apparatus according to any one of the preceding claims, wherein the second source is an annular light source. Apparatus according to any one of the preceding claims, wherein the second source comprises at least one LED (120). The device of any one of the preceding claims, wherein the second source comprises between 50 and 100 LEDs (120). Device according to one of Claims 7 - 8th , wherein the LEDs are mutually equidistant. Device after Claim 7 wherein the second source comprises a plurality of LEDs arranged in an annular pattern (121). Device after Claim 7 wherein the second source comprises a plurality of LEDs arranged in two spaced-apart annular patterns (121, 122). Device according to one of the preceding claims, wherein the opto-electrical module comprises silicon and / or amorphous silicon. An apparatus according to any one of the preceding claims, wherein the optoelectronic module represents the bottom plate of a screen, the bottom plate comprising a plurality of thin film transistors (103), a plurality of electrodes (104), and a plurality of capacitors (105). Device after Claim 13 , where the plurality is at least one million. Device according to one of the preceding claims, wherein the second source is arranged above the optoelectronic module. Device according to one of the preceding claims, wherein the second source is arranged below the optoelectronic module. Device according to one of the preceding claims, wherein the optoelectronic module has at least one transistor (103). The device of any one of the preceding claims, further comprising a test unit (18) having contacts for electrically contacting the device with contact elements mounted on the optoelectronic module. Device after Claim 18 , further comprising a control unit of the second source for turning on and off the second source and / or for synchronizing the second source with an electrical signal applied to the test unit. Device after Claim 18 or 19 , further comprising an electrical circuit for generating electrical signals at the test unit. Apparatus according to any one of the preceding claims, wherein the first source is a particle beam source for generating a particle beam. Apparatus according to any one of the preceding claims, wherein the first source is an electron beam source for generating an electron beam. Device according to one of the preceding claims, wherein the second source is arranged under a stage (9) for supporting the optoelectronic module. Device according to one of the preceding claims, wherein the optoelectronic module is a part of an LCD. Device according to one of the preceding claims, wherein the optoelectronic module is a screen base plate. Device according to one of the preceding claims, wherein the device comprises at least two first sources, at least two second sources, and at least two detectors. Device according to one of the preceding claims, wherein the second source provides light from the visible spectrum, in particular red light. Apparatus according to any one of the preceding claims, further comprising an electrical measuring circuit for measuring the voltage stored in pixels of the optical module, the electrical measuring circuit being connected to the contact elements of the optoelectronic module. Device according to one of the preceding claims, further comprising a stage (9) for holding the optoelectronic module. Device according to one of the preceding claims, wherein the optoelectronic module has no liquid crystals. Apparatus according to any one of the preceding claims, wherein the beam of the first source is for generating secondary particles measured by the detector. A method of testing an optoelectronic module, comprising: a. Illuminating the opto-electronic module for a lighting duration between 100 μs and 0.5 s; b. Directing an electromagnetic beam or particle beam, wherein the illumination is completed before judging; and c. Detecting defects of silicon and / or amorphous silicon in the optoelectronic module. Method for testing an optoelectronic module according to Claim 32 in which the illumination of the optoelectronic module takes place with electromagnetic radiation having a wavelength of at most 800 nm. Method according to Claim 32 or 33 wherein the electromagnetic beam or particle beam is generated from a first source and the illumination is performed by means of a second source. Method according to one of Claims 32 - 34 The method is carried out in a darkroom. Method according to one of Claims 32 - 35 , further comprising applying a voltage to the optoelectronic module. Method according to one of Claims 32 - 36 wherein the optoelectronic module comprises thin-film transistors, and the detection takes place only within a period of time in which the voltage at defect-free thin-film transistors of the optoelectronic module has dropped by not more than 20% of the previously applied voltage. Method according to one of Claims 32 - 37 , wherein the optoelectronic module comprises thin-film transistors, and the detection takes place only within a period in which the voltage at defect-free thin-film transistors of the optoelectronic module has dropped by not more than 40% of the previously applied voltage. Method according to one of Claims 32 - 38 , wherein the electromagnetic beam or particle beam is directed to the optoelectronic module. Method according to one of Claims 32 - 39 wherein the electromagnetic beam or particle beam is directed to a detector unit. Method according to one of Claims 32 - 40 , further comprising the step of measuring the voltage across at least one pixel of the optoelectronic module. Method according to one of Claims 32 - 41 , further comprising the step of measuring the voltage across a plurality of pixels of the optoelectronic module. Method according to Claim 42 wherein an average voltage value is calculated based on the measurement results of the voltage at the plurality of pixels, and the measurement of the voltage of each pixel is compared with the average voltage value. Method according to one of Claims 41 - 43 wherein a pixel is classified as defective if the measured voltage value deviates more than a limiting percentage from the average voltage value. Method according to one of Claims 32 - 44 wherein the illumination takes place before and during the straightening of the beam and / or before and during the detection. Method according to one of Claims 32 - 45 in which the illumination takes place for a time interval between 0.01 s and 0.1 s. Method according to one of Claims 32 - 46 wherein the illumination takes place only for a time interval in which the voltage across a defect-free thin-film transistor of the optoelectronic module has dropped by no more than 10% of the previously applied voltage. Method according to one of Claims 32 - 47 wherein the illumination occurs only for a time interval in which the voltage across a defect-free thin film transistor of the optoelectronic module has dropped by no more than 20% of the previously applied voltage. Method according to one of Claims 32 - 48 , further comprising the step of placing the optoelectronic module on a stage. Method according to one of Claims 32 - 49 wherein the voltage is applied to contact elements of the opto-electrical module for a period of time. Method according to one of Claims 32 - 50 wherein illuminating the optoelectronic module precedes directing a beam. Method according to one of Claims 32 - 51 wherein the illumination is performed such that the illumination arriving within a region comprising a plurality of pixels of the optoelectronic module is substantially homogeneous. Method according to one of Claims 32 - 52 wherein a drive is performed by the application of a voltage to the optoelectronic module. Device after Claim 53 , wherein there is a waiting time between driving and detecting. Method according to one of Claims 32 - 54 wherein the illumination takes place such that the illuminance on the optoelectronic module is about 500-1,500 lx. Method according to one of Claims 32 - 55 wherein detecting comprises measuring secondary electrons generated by an electron beam on the optoelectronic module. Method according to one of Claims 32 - 56 wherein the testing takes place in a vacuum chamber.

Images